Royal College of Surgeons in Ireland [email protected]Development of a Nanomedicine loaded Hydrogel for Sustained Delivery of Development of a Nanomedicine loaded Hydrogel for Sustained Delivery of an Angiogenic Growth Factor to the Ischaemic Myocardium an Angiogenic Growth Factor to the Ischaemic Myocardium AUTHOR(S) Joanne O'Dwyer, Robert Murphy, Eimear B. Dolan, Lenka Kovarova, Vladimir Velebny, Andreas Heise, Garry P. Duffy, Sally-Ann Cryan CITATION O'Dwyer, Joanne; Murphy, Robert; B. Dolan, Eimear; Kovarova, Lenka; Velebny, Vladimir; Heise, Andreas; et al. (2020): Development of a Nanomedicine loaded Hydrogel for Sustained Delivery of an Angiogenic Growth Factor to the Ischaemic Myocardium. figshare. Journal contribution. https://hdl.handle.net/10779/rcsi.12000864.v1 HANDLE 10779/rcsi.12000864.v1 LICENCE CC BY-NC-SA 4.0 This work is made available under the above open licence by RCSI and has been printed from https://repository.rcsi.com. For more information please contact [email protected]URL https://repository.rcsi.com/articles/Development_of_a_Nanomedicine_loaded_Hydrogel_for_Sustained_Deliver y_of_an_Angiogenic_Growth_Factor_to_the_Ischaemic_Myocardium/12000864/1
44
Embed
Development of a Nanomedicine loaded Hydrogel for ...€¦ · Development of a Nanomedicine loaded Hydrogel for Sustained Delivery of an Angiogenic Growth Factor to the Ischaemic
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Development of a Nanomedicine loaded Hydrogel for Sustained Delivery ofDevelopment of a Nanomedicine loaded Hydrogel for Sustained Delivery ofan Angiogenic Growth Factor to the Ischaemic Myocardiuman Angiogenic Growth Factor to the Ischaemic Myocardium
AUTHOR(S)
Joanne O'Dwyer, Robert Murphy, Eimear B. Dolan, Lenka Kovarova, Vladimir Velebny, Andreas Heise, GarryP. Duffy, Sally-Ann Cryan
CITATION
O'Dwyer, Joanne; Murphy, Robert; B. Dolan, Eimear; Kovarova, Lenka; Velebny, Vladimir; Heise, Andreas; etal. (2020): Development of a Nanomedicine loaded Hydrogel for Sustained Delivery of an Angiogenic GrowthFactor to the Ischaemic Myocardium. figshare. Journal contribution.https://hdl.handle.net/10779/rcsi.12000864.v1
HANDLE
10779/rcsi.12000864.v1
LICENCE
CC BY-NC-SA 4.0
This work is made available under the above open licence by RCSI and has been printed fromhttps://repository.rcsi.com. For more information please contact [email protected]
10 mM, Foetal Bovine Serum (FBS) 2% and Penicillin/Streptomycin 1%). VEGF was used at a
concentration of 5 ng/ml for feeding the cells during expansion but was removed, in all cases
for cell studies. Biocompatibility testing was carried out initially and the methods and results
are included in the supplementary material.
Tubule formation-Matrigel® assay
In vitro microvessel/tubule formation assays can be used as an indicator of potential in vivo
angiogenic activity [29-31]. 120 μl of growth factor reduced, phenol red free Matrigel® was
placed in each well of a 48 well plate and incubated at 37°C for 30 minutes. HUVECs were
seeded on top of the Matrigel® at a density of 3 x 104 cells per well in a spiral pattern [32].
Control groups consisted of endothelial growth medium with no VEGF or endothelial growth
medium with 50 ng/ml non-encapsulated VEGF. Three nanomedicine treatment groups were
used: L-PGA-VEGF 30:1, star-PGA-VEGF 30:1 and star-PGA-VEGF 50:1. Each nanomedicine
formulation contained 50 ng VEGF. Nanomedicines were added directly to the culture
medium. Tubule formation was assessed by taking five photographs of each well using a Leica
Microscope (Leica Microsystems, Switzerland) at 10X magnification at 6, 12 and 24 hours.
Total tubule length was determined by finding the sum of the tubule lengths per well using
ImageJ software (National Institutes of Health).
Cell migration-Scratch assay
HUVECs were seeded on a 24 well plate at a seeding density of 3 x 104 cells per well and fed
with complete endothelial medium until a confluent monolayer of cells was present in the
plate. At this point medium was removed from all wells and a P200 pipette tip was used to
scratch a vertical line down through the cell monolayer thus removing the cells in this location.
Each of the wells was washed three times with 500 µl PBS to remove the detached cells.
Treatments and controls were then added to the wells. FBS was removed from the medium
for this experiment to reduce the potential for cell proliferation interfering with results. Wells
were imaged at 5X magnification at time zero and the position of the image marked on the
plate to ensure consistency of imaging throughout the experiment. Images were taken at 0,
6, 12 and 24 hours. Gap distance was measured in three locations on the image using ImageJ
software.
Fabrication and characterisation of PGA-VEGF nanomedicine-loaded hyaluronic acid
hydrogels
Tyramine-modified hyaluronic acid (HA-TA) (250-350 kDa) was kindly provided by Contipro (Czech
Republic). Hydrogel formation involved dissolving the HA-TA in PBS at the relevant concentration,
taking account of the volume of nanoparticles to be added. The resulting dispersion was placed
on a roller-plate overnight to ensure complete wetting of the freeze-dried HA-TA powder.
Following this the HA-TA dispersion was divided into two separate Falcon tubes. The final
concentration was adjusted by addition of the nanoparticle dispersion to one tube and PBS to the
other to ensure both contained the same concentration of HA-TA in either PBS alone or a
PBS/nanoparticle containing solution. Horseradish peroxidase (HRP) was added to one Falcon
tube at a concentration of 0.24 U/ml and Hydrogen peroxide (H2O2)
was added to the other tube at 0.88 µmol/ml. These agents acted as the crosslinkers required
for gelation to occur. It was imperative to keep the crosslinkers separated until gelation was
required, thus the dispersions from each Falcon tube were drawn up into two separate 1 ml
syringes. These were then attached to a benchtop hydrogel mixer (BHM) (Fig. 1(b)) (Contipro,
Czech Republic) which contained a static mixer to facilitate correct interaction of the
components on injection to allow gelation. This system is shown in Fig. 1.
(a) (b)
Fig. 1(a) System used to formulate the nanomedicine-loaded hyaluronic acid hydrogel, showing addition of Horseradish peroxidase and PGA-VEGF nanomedicines to one syringe and hydrogen peroxide to the other syringe. Preparations then pass through the benchtop hydrogel mixer device (b) where the static mixer produces homongeous mixing of the two components to form a hydrogel.
Rheological testing
Rheology was performed on an AR-1000 cone and plate rheometer (TA Instruments, USA).
Time sweeps were performed at 21°C, under a shear stress of 5 Pa and frequency of 1 Hz for
thirty minutes.
Hydrogel homogeneity
Assessment of particle distribution in the hydrogel was achieved by fluorescently tagging the
star-PGA polypeptide using an Alexa Fluor™ 488 Microscale Protein Labeling Kit. Tagging was
performed using the kit as per the manufacturer’s guidelines. The final product was
centrifuged to remove any residual unbound tag and a standard curve was produced to
facilitate concentration determination. This fluorescent tag was capable of binding to a
primary amine group of the PGA polypeptide, without interfering with VEGF binding. Star-
PGA-VEGF 50:1 nanomedicines were then formulated using this fluorescently tagged PGA and
incorporated into the hydrogel as previously described. A confocal microscope (Carl Zeiss,
Germany) was used to visualise the dispersion of nanoparticles in the hydrogel.
Analysis of VEGF release and bioactivity from the nanomedicine-loaded hydrogel
A 200 µl cylinder of gel was placed inside a 1 ml cellulose ester membrane Spectra/Por® Float-A-
Lyzer® G2 device with a MWCO of 300 kDa and a further 200 µl of PBS was placed on top of the
hydrogel sample. The release study then proceeded as described in the ‘In vitro drug release
studies’ section described above, with VEGF detection again being performed via
ELISA. To ensure the bioactivity of the released VEGF, the release supernatant from all time
points up to day 14 was pooled, concentrated using an Amicon® Ultra 15 ml centrifugal filter
with a molecular weight cut-off of 3 kDa and the concentrated solution was applied to both a
Matrigel® and a scratch assay which again were carried out as described above.
Statistical analysis
All statistical tests were performed using GraphPad Prism v5 (GraphPad Software Inc., CA,
USA). Mean and standard error of the mean are presented on all graphs. A one-way ANOVA
followed by Bonferroni post-hoc test was used to analyse the data obtained from the
Matrigel® and scratch assays. Significance was determined as p<0.05. Three repeats were
performed for all experiments.
Results
Synthesis of glutamic acid–based polypeptides
L-PGA contained 200 glutamic acid residues per molecule (Mn: 26 kDa) (Fig. 2(a)). The star-
shaped PGA had a polypropyleneimine (PPI) core and eight arms, each with 40 glutamic acid
residues (theoretical Mn: 42 kDa; estimated isoelectric point: 4.1). The structure of this star-
PGA is shown in Fig. 2(b). Gel permeation chromatograms for both L-poly-γ-benzyl-L-
glutamate (PBLG) and star PBLG, the precursors to L-PGA and star-PGA are shown in Online
Resource 1.
(a) (b)
Fig. 2(a) The linear PGA and (b) star-PGA polypeptides synthesised which will be used for
nanomedicine fabrication in this work.
Characterisation of PGA-VEGF formulations
Particle formulation
Particle dispersions were formed by adding star-PGA or L-PGA to an eppendorf followed by the
requisite amount of VEGF and PBS to a final volume of 50 µl. Seven different molar ratios of star-
PGA:VEGF were initially trialled based on a star-PGA molecular weight of 42 kDa and a VEGF
molecular weight of 42 kDa. Four complexation procedures were used: 5 minutes complexation
at room temperature, 30 minutes complexation at room temperature (~21°C) and either 5 or 30
minutes complexation in the fridge (2-8°C). The results from these initial studies are shown in
Online Resource 2. Based on the particle sizes shown in Online Resource 2, 5 minutes
complexation at room temperature was chosen as the optimal complexation technique because
this procedure produced nano-sized particles while also being appropriate for translation of
manufacture to a larger scale. L-PGA-VEGF formulations were complexed
for 5 minutes at room temperature. As stated previously, in this paper we will focus on three
formulations L-PGA-VEGF 30:1, star-PGA-VEGF 30:1 and starPGA-VEGF 50:1, the rationale for
choosing these formulations is discussed in Online Resource 2.
Particle size and Zeta potential
As shown in Table 1, DLS indicated the formation of nano-sized particles for L-PGA-VEGF 30:1
and star-PGA-VEGF 30:1 and 50:1 formulations. Of these three formulations L-PGA-VEGF was
the largest with a Z-average size of 656.7 nm. The star-PGA-VEGF 30:1 and 50:1 formulations
were very similar in size with Z-averages of 444.5 nm and 415.5 nm respectively. The Zeta
potential data demonstrated a negative surface charge for all formulations. This charge was
very close to neutral for the star-PGA-VEGF 30:1 and 50:1 formulations at -2.3 meV and -3.6
meV respectively. The Zeta potential of the L-PGA-VEGF 30:1 formulation was slightly more
negative than that of the star-PGA-VEGF formulations at -7.7 meV. Particle size
measurements on the Nanosight, 203 nm for star-PGA-VEGF 30:1 and 196.4 nm for star-PGA-
VEGF 50:1, indicated smaller sizes than those observed with DLS for star-PGA-VEGF
nanoparticles. However, similar to the DLS data the size of star-PGA-VEGF 30:1 was similar to
that of star-PGA-VEGF 50:1. L-PGA-VEGF formulations could not be observed on the
Nanosight. Encapsulation efficiency for all formulations was greater than 99.99% w/w
indicating almost complete encapsulation of the VEGF. Loading capacity varied from 2% w/w
to 3.33% w/w (50 ng/1.5 µg to 50 ng/2.5 µg) depending on the molar ratio of PGA to VEGF
used.
Table 1 Physicochemical characterisation of PGA-VEGF formulations (n=3). No particles were detected for L-PGA-VEGF 30:1.
Formulation Z-average Zeta Polydispersity Average size Encapsulation Loading size (nm) Potential Index Nanotracking
In vitro release of VEGF from PGA-VEGF nanomedicines
Following confirmation of nanomedicine formation and VEGF encapsulation, the rate of VEGF
release from the nanomedicines was determined. Results are shown as the cumulative
percentage of VEGF originally encapsulated released up to and including the relevant time
point. As shown in Fig. 3 L-PGA-VEGF formulations exhibited more rapid release with 30% of
the loaded VEGF released at day 7, compared to 17% and 18% release from the star-PGA-
VEGF 30:1 and 50:1 formulations respectively. No release was detected from this L-PGA-VEGF
formulation after day 7. In contrast the two star-PGA-VEGF formulations released VEGF for 28
days, with no release detected thereafter. There was no significant difference between the
two star-PGA formulations, in terms of release at any time point.
Rel
e
ased
40
L-PGA-VEGF 30:1
30
Star-PGA-VEGF 30:1
VE
GF
Star-PGA-VEGF 50:1
20
Cum
ulat
ive
10
% 0
0 10 20 30
Time (Days)
Fig. 3 Cumulative percentage VEGF released from L-PGA-VEGF or star-PGA-VEGF formulations over 28 days. The star-PGA-VEGF formulations exhibited more sustained VEGF release than the L-PGA-VEGF formulation. All formulations contained 50 ng VEGF. n=3.
Bioactivity of PGA-VEGF nanomedicines: tubule formation
The bioactivity of the PGA-VEGF nanomedicines was next assessed by investigating their
ability to induce microvessel/tubule formation in vitro. Total tubule length was measured here
as this was deemed to be the most appropriate parameter for this application. Linear and
star-PGA-VEGF (50 ng VEGF) nanomedicine formulations are compared to controls of
untreated cells (cells alone) and cells exposed to fresh, non-encapsulated VEGF at a
comparative dose of 50 ng. Fig. 4(a) shows that, overall, much better interlinked microvessel
networks are present in groups treated with VEGF. As presented in Fig. 4(b) star-PGA-VEGF
30:1 and 50:1 formulations were capable of inducing the same degree of tubule formation at
6 hours as non-encapsulated VEGF and all treatment groups significantly increased total
tubule length compared to cells alone. At 12 hours, star-PGA-VEGF 30:1 and 50:1 were again
significantly better than cells alone. In contrast, the L-PGA-VEGF nanomedicines were not
significantly better than cells alone at 6 hours. The L-PGA formulation did exhibit a significant
increase in tubule length over cells alone at 12 hours, although less than the increase
observed for the star-PGA-equivalents. At 24 hours, tubule regression is occurring in all groups
and tubule length is reduced. This is considered to be the typical behaviour of such structures
in this assay [32].
(a) (i) (ii) (iii)
100 µm 100 µm 100 µm
Cells Alone Non-encapsulated VEGF L-PGA-VEGF 30:1
(iv) (v)
100 µm 100 µm
Star-PGA-VEGF 30:1 Star-PGA-VEGF 50:1 (b)
To
tal T
ub
ule
Len
gth
(µ
m)
60000 **
Cells Alone ** **
Non-encapsulated VEGF
**
***
** L-PGA-VEGF 30:1
40000 * Star-PGA-VEGF 30:1
Star-PGA-VEGF 50:1
20000
0 6 Hours 12 Hours 24 Hours
Fig. 4 Bioactivity of VEGF released from PGA-VEGF nanoparticles (a) Microscopy images of tubule formation following 12 hours of culture. All groups display better tubule formation than the cells alone group not exposed to VEGF. (b) Quantification of tubule lengths at 6, 12 and 24 hours, confirming significantly better tubule lengths with all VEGF containing groups at 12 hours. All formulations contained 50 ng VEGF. n=3; *p<0.05, **p<0.01, ***p<0.001.
Bioactivity of PGA-VEGF nanomedicines-cell migration
Cell migration is also a key process in in vivo angiogenesis and therefore the efficacy of the
VEGF formulations was investigated using a scratch assay. 6 hours following gap formation
and treatment addition, non-encapsulated VEGF treatment had led to a significant reduction
in the gap width, but no significant reduction was seen for the other treatment groups (Fig.
5(b)). At 12 hours the gap width was 23% of the original width in the non-encapsulated VEGF
group and 20% of the original width in the star-PGA-VEGF 50:1 nanomedicine treated group.
Both of these groups, non-encapsulated VEGF and star-PGA-VEGF 50:1 had significantly
reduced the gap width compared to cells alone (gap width 53% of original width). Both non-
encapsulated VEGF and star-PGA-VEGF 50:1 effected closure of the gap by 24 hours while
cells alone, L-PGA-VEGF 30:1 or star-PGA-VEGF 30:1 nanomedicines did not achieve gap
closure at 24 hours.
(a)
(i) (ii) (iii)
100 µm 100 µm 100 µm
Cells Alone Non-encapsulated VEGF L-PGA-VEGF 30:1
(iv) (v)
100 µm 100 µm
(b) Star-PGA-VEGF 30:1 Star-PGA-VEGF 50:1
Gap
wid
th a
s a
% o
f o
rig
inal w
idth
100
80
60
40
20
0
6 Hours
Cells Alone Non-encapsulated VEGF
L-PGA-VEGF 30:1 Star-PGA-VEGF 30:1
Star-PGA-VEGF 50:1 **
**
* * 12 Hours 24 Hours
Fig. 5 Cell migration assessed via gap closure achieved on a scratch assay (a) Microscopy images of the gap area 12 hours following addition of treatments. (b) Quantification of gap closure showing the star-PGA-VEGF 50:1 formulation closing the formed gap as efficiently as the non-encapsulated VEGF. Where VEGF is present the concentration is 50 ng/ml. n=3; *p<0.05, **p<0.01.
Incorporation of star-PGA-VEGF 50:1 nanomedicines into a HA-TA hydrogel
Star-PGA-VEGF 50:1 nanomedicines, which had produced significant results in both Matrigel®
and scratch assays, were incorporated into a HA-TA polymer dispersion as described.
Rheology was performed on the formulation following transit through the benchtop hydrogel
mixer (BHM), as shown in Fig. 1, to assess the effect of nanomedicine incorporation on
hydrogel formation. Hydrogel formation is indicated when the storage modulus (G’) exceeds
the loss modulus (G’’). Fig. 6(a) shows that despite the addition of the nanomedicines to the
HA-TA dispersion, gelation still occurred and equilibration of storage modulus was achieved
more rapidly than with the HA-TA formulation alone. Overall, storage modulus was reduced
in the presence of the star-PGA-VEGF 50:1 nanomedicines. Fig. 6(b) shows the dispersion of
the fluorescently tagged star-PGA-VEGF nanomedicines in a circular section of the hydrogel,
confirming their ability to travel through the BHM and their homogenous distribution in the
resulting hydrogel.
(a) 1500 HA-TA G'
(b) 1%
1% HA-TA G''
G' o
r G
'' (
Pa
)
1000 1% HA-TA + nanoparticles G'
1% HA-TA + nanoparticles G''
500
0 0 500 1000 1500 2000
Time (s)
Fig. 6 (a) Rheogram obtained from a thirty minute time sweep showing storage (G’) and loss (G’’) modulus of the HA-TA gel with or without star-PGA-VEGF 50:1 nanomedicines. Gel formation occurs even in the presence of the nanomedicines. (b) Dispersion of fluorescently tagged star-PGA-VEGF 50:1 nanomedicines in the HA-TA hydrogel. VEGF concentration is 50 ng per 200 µl hydrogel portion.
VEGF release and bioactivity of the star-PGA-VEGF nano-in gel system
Free VEGF and PGA-VEGF nanomedicines were both loaded into separate HA-TA hydrogels
and VEGF release and the bioactivity of released VEGF determined. Free VEGF was rapidly
released from the HA-TA hydrogel with 17.3% of the loaded VEGF released by 24 hours (Fig.
7(a)). No further VEGF release was detected from the hydrogel system for this formulation.
Where VEGF was incorporated into the gel as a PGA-VEGF nanomedicine, only 2.86% of the
VEGF was released at 24 hours, with release continuing to be detected for up to 35 days.
Loading star-PGA-VEGF 50:1 nanomedicines in the HA-TA hydrogel increased VEGF recovery
to 45% compared to a recovery of just 17% when free VEGF was loaded into the HA-TA
hydrogels. At the end of the experiment, all hydrogels were degraded with hyaluronidase and
the resulting liquid was applied to an ELISA. No VEGF was found in these degraded hydrogels
(data not shown).
To confirm that the bioactivity of the released VEGF was not affected by the hydrogel fabrication
process Matrigel® and scratch assays were carried out using the pooled, concentrated VEGF
collected from the release supernatants. Fig. 7 (b) and (c) show that VEGF released from both gel
systems was capable of producing tubule lengths analogous to those produced by the control
treatment of a comparative dose of non-encapsulated, fresh VEGF. Furthermore, VEGF released
from the PGA-VEGF-HA-TA system significantly improved total tubule length at the 6 hour
timepoint compared to cells alone. This retained bioactivity of
released VEGF was also evidenced in the scratch assay where gap closure produced by the
VEGF released from the star-PGA-VEGF 50:1 loaded HA-TA gel was similar to that produced
by the same dose of fresh, free VEGF. There was no significant difference in effect between (a)
the VEGF released from the free VEGF loaded into HA-TA and star-PGA-VEGF loaded HA-TA.
(b)
(c)
Fig. 7 (a) VEGF released from HA-TA hydrogels either with or without initial incorporation into a nanoparticle. Nanoparticle encapsulated VEGF exhibited more sustained release from the hydrogel than non-encapsulated VEGF. (b) Total length of tubules recorded in response to VEGF released from hydrogels with non-encapsulated VEGF used as a bioactivity control. The total tubule length produced in response to the VEGF released from HA-TA hydrogels is in the
same range as that induced by fresh, non-encapsulated VEGF. (c) Gap closure on a scratch assay again indicating that VEGF released from HA-TA hydrogels is bioactive. Fresh, free VEGF: 10 ng/ml, Free VEGF-HA-TA: 8.65 ng/ml, PGA-VEGF-HA-TA: 11.5 ng/ml. n=3; *p<0.05, **p<0.01, ***p<0.001.
Discussion
One of the main issues hindering the exploitation of growth factors as therapeutics for
biomedical applications is their rapid degradation in vivo [33, 34]. The angiogenic growth
factor VEGF has been proposed to be useful for applications where improved blood supply to
a target tissue is required. Herein a novel approach to improve the stability and control the
release of VEGF, using an oppositely charged polypeptide-based carrier, was investigated. This
carrier is biocompatible and employs a non-destructive manufacturing process, amenable to
scale-up for loading the VEGF. These VEGF nanomedicines were loaded into a hyaluronic acid
hydrogel similar to one which has previously shown promise for its ability to improve
outcomes post-MI in animal models [35].
Negatively charged PGA polypeptides with either a star-shaped or linear architecture were
synthesised to bind electrostatically to the positively charged site on VEGF. Neither of these
polypeptides have, to the best of our knowledge, previously been investigated for their ability
to complex with VEGF. Both the linear and star-shaped architectures were trialled as such
differences in polymer architecture have previously been shown to alter binding capabilities
with therapeutics and in vivo half-life of nanomedicines [36]. Two star-PGA based
formulations were used to assess if different molar ratios affected particle formation and/or
activity and an L-PGA-VEGF 30:1 nanomedicine formulation was used as a direct comparator
to the star-PGA-VEGF 30:1 formulation again to assess differences in particle characteristics
when formed with different polypeptide architectures.
Nanoparticles are an attractive modality for controlled release in cardiac applications with a lower
potential for embolization than microparticles [14, 15]. No optimal criteria for nanoparticle
delivery to the heart have been established but previous publications have had some success in
in vivo studies with particles ranging from 115 nm to 400 nm, which efficiently delivered the
loaded growth factor and did not produce any adverse events [16, 18, 20].
Both L-PGA-VEGF and star-PGA-VEGF formulations produced nanomedicines with a low PDI
on DLS measurements with slightly negative surface charges which was expected given the
excess of PGA in the formulations. Star-PGA-VEGF 30:1 and 50:1 formulations were
approximately 400 nm when measured using DLS but only 200 nm when measured by NTA.
This phenomenon is not unusual and has been reported in the literature as being due to the
differing modes of operation of the two instruments [37, 38]. The size of the star-PGA-VEGF
30:1 and 50:1 nanomedicines exceeds that of 50 nm observed by Yan et al formulating insulin
nanoparticles with a star PLL-Poly(ethylene glycol) (PEG) polypeptide using a similar
procedure [22]. This disparity may be due to the large difference in size between VEGF (42
kDa) and insulin (5 kDa). Star PLL-based formulations used to encapsulate plasmid DNA have
previously been shown to be in the range of 140 nm [39].
L-PGA-VEGF nanomedicines were not evident when using the NTA. This may indicate inferior
VEGF binding in the L-PGA-VEGF formulations. It has previously been shown that in vivo the half-
life of L-PGA is 13 times shorter than a star-shaped comparator and this may have an impact on
particle stability even in vitro [36]. Furthermore, nanomedicines formed from star
polypeptides may have a more globular shape than their linear counterparts which may also
explain why detection using NTA which is optimised for globular systems is more difficult [36].
Kita et al. have suggested that optimising encapsulation efficiency should be the top priority
in the development of drug delivery systems and an encapsulation efficiency of >99.9% was
achieved for all PGA-VEGF formulations tested, both linear and star-shaped [41].
Encapsulation efficiencies for VEGF-loaded nanoparticles reported in the literature vary
greatly from 53.5% for PLGA-based nanoparticles synthesised by Oduk et al, to 76% as
reported by des Rieux et al. for dextran-chitosan-based nanoparticles [18, 20]. The
encapsulation efficiency obtained in this study for PGA-VEGF nanoparticles mirrors that
obtained by Yan et al. for insulin encapsulation in their star PLL-PEG system which was
reported as “almost 100%”[22]. The maximum loading capacity feasible in this system was
3.33% w/w (50 ng/1.5 µg). Again this was similar to the star PLL-insulin nanoparticles
formulated by Yan et al, who reported a maximum loading capacity of 4.9% [22]. Golub et al.,
Oduk et al. and des Rieux et al. reported loading capacities of 5.3%, 0.01% and 0.5%
respectively when loading VEGF into more traditional PLGA (5.3 and 0.01%) or dextran-
chitosan-based (0.5%) nanoparticles [16, 18, 20]. The high encapsulation efficiency and ease
of loading is a significant advantage of this PGA formulation, preventing loss of expensive
therapeutic protein during nanoparticle fabrication, a major translational hurdle.
Prolonged release of VEGF is required to facilitate optimal vessel growth [13, 41]. Star-PGA-
VEGF 50:1 nanomedicines released 72% of the total released VEGF by day 7 with further
release observed until day 28. In contrast, VEGF release from the L-PGA-VEGF 30:1
formulation was more rapid. There are many possible reasons for the more rapid release
observed with the L-PGA-VEGF 30:1 nanomedicine. Firstly, as previously stated the L-PGA has
a much shorter half-life in vivo than its star-shaped counterpart [36]. Similarly, Byrne et al.
found that even in the presence of the enzymes thermolysin and chymotrypsin release of
rhodamine B from L-PGA was more rapid and complete than that observed with an 8 armed
star-PGA similar to the one used in this work [23].
Silva et al. have previously observed in vivo VEGF release from an alginate gel containing 250
ng VEGF/ml of hydrogel of 60% over the first seven days followed by release of a further 20%
up to day 28. The released VEGF significantly improved murine hindlimb perfusion [13]. Thus,
the in vitro release profile from the star-PGA-VEGF nanomedicines, with most release
occurring over the first seven days followed by release of small amounts of VEGF up to day 28
could potentially improve vessel formation in vivo. Less than 30% of the loaded VEGF was
recovered overall during release studies. In light of a reported 90 minute half-life at 37°C this
is neither surprising nor uncommon and has been reported previously in the literature [42].
Furthermore studies where such low recoveries of VEGF have been noted have gone on to be
efficacious in inducing angiogenesis in animal models with no apparent associated toxicity
[18, 32].
Star-PGA-VEGF 30:1 and 50:1 nanomedicines were capable of inducing significantly better
tubule lengths at both 6 and 12 hours compared to cells alone. Furthermore, the increase in
tubule length observed at 6 hours for both star-PGA-VEGF nanomedicines is as statistically
significant as that seen for non-encapsulated VEGF at 6 hours. The star-PGA-VEGF
nanomedicines have only released approximately 3.5 ng (30:1) and 1 ng (50:1) VEGF
respectively, in comparison to the 50 ng in the control group. This greater effect may be due
to increased potency of the VEGF released which, unlike the VEGF in the control group, is
possibly not being degraded and is protected inside the nanoparticle. Despite having released
approximately 5 ng of VEGF by 6 hours, the L-PGA-VEGF formulation has less significant effects
on tubule length with a significant increase in length only seen at 12 hours. This may be due
to the possible toxic effects of the L-PGA molecule itself (Online Resource S4). Overall the
results of this experiment indicate that the star-PGA-VEGF 30:1 and 50:1 formulations may be
superior to their L-PGA-VEGF 30:1 counterpart. Vessel sprouting is one of the main processes
involved in the initiation and progression of angiogenesis in vivo [30, 43]. Previously, VEGF
induced microvessel formation in vitro has translated to improvements in in vivo vessel
formation and density [17, 20].
As well as vessel formation assays, cell migration assays can also be of use in assessing the in
vivo angiogenic potential of formulations [43, 44]. The star-PGA-VEGF 50:1 nanomedicines
were capable of inducing significant migration producing gap closure at 24 hours. Neither the
L-PGA-VEGF 30:1 nor the star-PGA-VEGF 30:1 nanomedicines were capable of achieving gap
closure at 24 hours. Therefore, star-PGA-VEGF 50:1 nanomedicines were selected for further
development. Overall the results obtained from this scratch assay are in line with those
obtained by Anderson et al. with their Heparin-VEGF nanoparticles which achieved 60% gap
closure at 18 hours and these particles went on to improve vessel formation in an in vivo assay
[45].
The star-PGA-VEGF nanomedicines were then incorporated into a hydrogel in order to
effectively deliver and retain the nanomedicines at the site of action. Hydrogels have shown
promise in their own right for improving patient symptoms following an MI [27]. Hydrogels
which can be injected through a catheter system may also aid minimally invasive delivery of
the nanomedicines. Hyaluronic acid is a molecule which has been in clinical use for over thirty
years and its safety has been well established in both the pharmaceutical and cosmetic sectors
[28]. Chemical crosslinking mechanisms mean that incorporation of cargo during the
formulation process of these hydrogels is possible.
Star-PGA-VEGF 50:1 nanomedicines were incorporated into a HA-TA hydrogel. Rheology
confirmed that hydrogel formation was still possible following inclusion of the PGA-VEGF
nanomedicines in the hyaluronic acid dispersion. A higher G’ was achieved in the absence of
the star-PGA-VEGF 50:1 nanomedicines indicating some form of interaction between the PGA
and the crosslinkers which are then impeded from forming the normal links within the
hydrogel, lowering the storage modulus. The storage modulus observed here is within the
range of moduli previously identified as suitable for endocardial delivery of hydrogels, with
Algisyl® having a reported storage modulus of 3-5 kPa [27]. Injections of hydrogels into the
heart have previously consisted of multiple injections around the site of damage [27]. Fig. 6(b)
indicates that the fluorescently tagged star-PGA-VEGF 50:1 nanomedicines are evenly
distributed in the hydrogel which confirms uniform mixing of the precursor dispersions.
VEGF release, both free and within star-PGA nanomedicines, from the HA-TA hydrogels was
investigated. Free VEGF loaded into HA-TA was released rapidly with 17% recovered within 24
hours. No further VEGF release was recorded thereafter. In contrast, when the star-PGA-VEGF
50:1 nanomedicines were integrated into the hydrogel VEGF release was detected for up to 35
days. Overall, VEGF recovery was also higher from this system with 42% quantified over the
release study period compared to 17% for free VEGF in the HA-TA system. Such differences in
VEGF release when incorporated into a double matrix system have previously been reported in
the literature. Des Rieux et al. reported 20% VEGF release over 14 days when
1 µg VEGF was incorporated into 500 µl Matrigel® compared to just 2% VEGF release when
dextran-chitosan-VEGF nanoparticles were loaded into the Matrigel® [18].
Bioactivity of the VEGF released from the hydrogel systems was determined using a scratch
assay and a Matrigel® assay. By day 14 the hydrogel containing non-encapsulated, ‘free’ VEGF
had released 8.65 ng VEGF, while that containing the star-PGA-VEGF 50:1 nanomedicines had
released 11.5 ng VEGF. Thus a dose of 10 ng free, fresh VEGF was chosen as an appropriate
control. The results of the Matrigel® assay (Fig. 7(b)) show that the VEGF released from both
preparations was capable of inducing increases in tubule length similar to those induced by
control VEGF. This result is confirmed on the scratch assay where gap closure was achieved
by 24 hours with VEGF released from PGA-VEGF-HA-TA. Overall the tubule formation
exhibited by all groups in Fig. 7(a) is less than that seen in Fig. 4. This is reasonable due to the
reduced VEGF dose used in this experiment to facilitate comparison with the amount of VEGF
released from the hydrogel.
Taken together these results suggest the development of an innovative sustained delivery
system for the angiogenic growth factor VEGF, which may overcome the current technical
barriers to effective growth factor delivery to improve post-MI angiogenesis and reduce
progression to heart failure.
Conclusion
In this paper we have successfully developed a sustained release nano-in gel system that
facilitates sustained release of VEGF for use in cardiac applications.
A method for the effective loading of linear and star-PGA polypeptides with VEGF was
developed which maintains the integrity of the growth factor and can be easily scaled-up.
Both PGA architecture and formulation parameters (molar ratios of PGA:VEGF) were found to
be critical in developing an effective controlled release and bioactive formulation for VEGF.
Star-PGA-VEGF nanomedicines formulated at a PGA-VEGF ratio of 50:1 were found to provide
sustained VEGF release while maintaining bioactivity post encapsulation. These PGA-VEGF
50:1 nanomedicines were successfully incorporated into a hyaluronic acid dispersion and gel
formation occurred following injection through the BHM. Finally we report the formation of
a bioactive, sustained release, double matrix system for the delivery of the angiogenic growth
factor VEGF. This PGA-VEGF-HA-TA system maintains VEGF bioactivity and has favourable
material and fabrication properties for scale-up, thus making it a prototype delivery platform
system for other growth factors of interest, advancing the delivery of these labile cargoes and
improving the lives of patients post-MI.
Acknowledgements:
The authors acknowledge the support of Brenton Cavanagh for assistance with imaging and
Matrigel® analysis. This study was undertaken as part of the Translational Research in
Nanomedical Devices (TREND) research project, School of Pharmacy, RCSI, facilitated via a
Science Foundation Ireland Investigators Program 13/IA/1840.
Conflict of Interest Statement:
References
1. RN P. The Circulatory System and Oxygen Transport. In: Sciences MCL, editor. Regulation of Tissue Oxygenation San Rafael (CA)2011 2. Alberts B, Johnson A, J L. Blood Vessels and Endothelial Cells. In: Science G, editor. Molecular Biology of the Cell. New York2002. 3. Buja LM, Vander Heide RS. Pathobiology of Ischemic Heart Disease: Past, Present and Future. Cardiovascular Pathology. 2016;25(3):214-20. 4. O'Brien FJ. Biomaterials & scaffolds for tissue engineering. Materials Today. 2011;14(3):88-95. 5. Duffy GP, McFadden TM, Byrne EM, Gill S, FarrelL E, O’Brien FJ. Towards in vitro vascularisation of Collagen-GAG scaffolds. European Cells and Materials. 2011;21:15-30. 6. Shah AM, Mann DL. In Search of New Therapeutic Targets and Strategies for Heart Failure: Recent Advances In Basic Science. Lancet. 2011;378(9792):704-12. 7. Cochain C, Channon KM, Silvestre J-S. Angiogenesis in the Infarcted Myocardium. Antioxidants & Redox Signaling. 2013;18(9):1100-13. 8. Adair T, Montani J. Angiogenesis. In: Sciences MCL, editor. Overview of Angiogenesis. San Rafael (CA)2010. 9. Ferrara N, Henzel WJ. Pituitary follicular cells secrete a novel heparin-binding growth factor specific for vascular endothelial cells. Biochemical and Biophysical Research Communications. 1989;161(2):851-8. 10. Thomas K. Vascular Endothelial Growth Factor, a potent and selective angiogenic agent. The Journal of Biological Chemistry. 1996;271(2):603-6. 11. Henry TD, Rocha-Singh K, Isner JM, Kereiakes DJ, Giordano FJ, Simons M, et al. Intracoronary administration of recombinant human vascular endothelial growth factor to patients with coronary artery disease. Am Heart J. 2001;142(5):872-80. 12. Henry TD. The VIVA Trial: Vascular Endothelial Growth Factor in Ischemia for Vascular Angiogenesis. Circulation. 2003;107(10):1359-65. 13. Silva EA, Mooney DJ. Effects of VEGF temporal and spatial presentation on angiogenesis. Biomaterials. 2010;31(6):1235-41. 14. Panyam J, Labhasetwar V. Biodegradable nanoparticles for drug and gene delivery to cells and tissue. Advanced Drug Delivery Reviews. 2003;55(3):329-47. 15. Davda J, Labhasetwar V. Characterization of nanoparticle uptake by endothelial cells. International Journal of Pharmaceutics. 2002;233(1):51-9. 16. Golub JS, Kim Y, Duvall CL, Bellamkonda RV, Gupta D, Lin AS, et al. Sustained VEGF delivery via PLGA nanoparticles promotes vascular growth. Am J Physiol Heart Circ Physiol. 2010;298(6):H1959- 65. 17. Xie J, Wang H, Wang Y, Ren F, Yi W, Zhao K, et al. Induction of angiogenesis by controlled delivery of vascular endothelial growth factor using nanoparticles. Cardiovasc Ther. 2013;31(3):e12-8.
18. des Rieux A, Ucakar B, Mupendwa BP, Colau D, Feron O, Carmeliet P, et al. 3D systems delivering VEGF to promote angiogenesis for tissue engineering. J Control Release. 2011;150(3):272-8. 19. Gu F, Amsden B, Neufeld R. Sustained delivery of vascular endothelial growth factor with alginate beads. J Control Release. 2004;96(3):463-72. 20. Oduk Y, Zhu W, Kannappan R, Zhao M, Borovjagin AV, Oparil S, et al. VEGF nanoparticles repair the heart after myocardial infarction. American Journal of Physiology-Heart and Circulatory Physiology. 2018;314(2):H278-H84. 21. Desai N. Challenges in Development of Nanoparticle-Based Therapeutics. The AAPS Journal. 2012;14(2):282-95. 22. Yan Y, Wei D, Li J, Zheng J, Shi G, Luo W, et al. A poly(l-lysine)-based hydrophilic star block co-polymer as a protein nanocarrier with facile encapsulation and pH-responsive release. Acta Biomaterialia. 2012;8(6):2113-20. 23. Byrne M, Thornton PD, Cryan S-A, Heise A. Star polypeptides by NCA polymerisation from dendritic initiators: synthesis and enzyme controlled payload release. Polymer Chemistry. 2012;3(10):2825. 24. Wang X, Wu X, Xing H, Zhang G, Shi Q, E L, et al. Porous Nanohydroxyapatite/Collagen Scaffolds Loading Insulin PLGA Particles for Restoration of Critical Size Bone Defect. ACS Applied Materials & Interfaces. 2017;9(13):11380-91. 25. Saludas L, Pascual-Gil S, Prosper F, Garbayo E, Blanco-Prieto M. Hydrogel based approaches for cardiac tissue engineering. Int J Pharm. 2017;523(2):454-75. 26. Van Vlierberghe S, Dubruel P, Schacht E. Biopolymer-based hydrogels as scaffolds for tissue engineering applications: a review. Biomacromolecules. 2011;12(5):1387-408. 27. Mann DL, Lee RJ, Coats AJ, Neagoe G, Dragomir D, Pusineri E, et al. One-year follow-up results from AUGMENT-HF: a multicentre randomized controlled clinical trial of the efficacy of left ventricular augmentation with Algisyl in the treatment of heart failure. Eur J Heart Fail. 2016;18(3):314-25. 28. Burdick JA, Prestwich GD. Hyaluronic acid hydrogels for biomedical applications. Adv Mater. 2011;23(12):H41-56. 29. Tahergorabi Z, Khazaei M. A Review on Angiogenesis and Its Assays. Iranian Journal of Basic Medical Sciences. 2012;15(6):1110-26. 30. Auerbach R, Lewis R, Shinners B, Kubai L, Akhtar N. Angiogenesis Assays: A Critical Overview. Clinical Chemistry. 2003;49(1):32-40. 31. Arnaoutova I, Kleinman HK. In vitro angiogenesis: endothelial cell tube formation on gelled basement membrane extract. Nat Protoc. 2010;5(4):628-35. 32. Quinlan E, Lopez-Noriega A, Thompson EM, Hibbitts A, Cryan SA, O'Brien FJ. Controlled release of vascular endothelial growth factor from spray-dried alginate microparticles in collagen-hydroxyapatite scaffolds for promoting vascularization and bone repair. J Tissue Eng Regen Med. 2015. 33. Lee K, Silva EA, Mooney DJ. Growth factor delivery-based tissue engineering: general approaches and a review of recent developments. J R Soc Interface. 2011;8(55):153-70. 34. Taraballi F, Pandolfi L, Powell S, Cabrera FJ, A T, Minardi S, et al. Strategic Approaches to Growth Factors Delivery for Regenerative Medicine
In: Panseri S, Taraballi F, Cunha C, editors. Biomimetic Approaches for Tissue Healing: OMICS Group; 2015. 35. Rodell CB, Lee ME, Wang H, Takebayashi S, Takayama T, Kawamura T, et al. Injectable Shear-Thinning Hydrogels for Minimally Invasive Delivery to Infarcted Myocardium to Limit Left Ventricular Remodeling. Circ Cardiovasc Interv. 2016;9(10). 36. Duro-Castano A, England RM, Razola D, Romero E, Oteo-Vives M, Morcillo MA, et al. Well-Defined Star-Shaped Polyglutamates with Improved Pharmacokinetic Profiles As Excellent Candidates for Biomedical Applications. Molecular Pharmaceutics. 2015;12(10):3639-49.
37. Anderson W, Kozak D, Coleman VA, Jamting AK, Trau M. A comparative study of submicron particle sizing platforms: accuracy, precision and resolution analysis of polydisperse particle size distributions. J Colloid Interface Sci. 2013;405:322-30. 38. Filipe V, Hawe A, Jiskoot W. Critical evaluation of Nanoparticle Tracking Analysis (NTA) by NanoSight for the measurement of nanoparticles and protein aggregates. Pharm Res. 2010;27(5):796-810. 39. Walsh DP, Murphy RD, Panarella A, Raftery RM, Cavanagh B, Simpson JC, et al. Bioinspired Star-Shaped Poly(l-lysine) Polypeptides: Efficient Polymeric Nanocarriers for the Delivery of DNA to Mesenchymal Stem Cells. Molecular Pharmaceutics. 2018. 40. Kita K, Dittrich C. Drug delivery vehicles with improved encapsulation efficiency: taking advantage of specific drug–carrier interactions. Expert Opinion on Drug Delivery. 2011;8(3):329-42. 41. Rui J, Dadsetan M, Runge MB, Spinner RJ, Yaszemski MJ, Windebank AJ, et al. Controlled release of vascular endothelial growth factor using poly-lactic-co-glycolic acid microspheres: In vitro characterization and application in polycaprolactone fumarate nerve conduits. Acta biomaterialia. 2012;8(2):511-8. 42. Kleinheinz J, Jung S, Wermker K, Fischer C, Joos U. Release kinetics of VEGF(165 )from a collagen matrix and structural matrix changes in a circulation model. Head & Face Medicine. 2010;6:17-. 43. Staton CA, Reed MW, Brown NJ. A critical analysis of current in vitro and in vivo angiogenesis assays. Int J Exp Pathol. 2009;90(3):195-221. 44. Liang CC, Park AY, Guan JL. In vitro scratch assay: a convenient and inexpensive method for analysis of cell migration in vitro. Nat Protoc. 2007;2(2):329-33. 45. Anderson SM, Siegman SN, Segura T. The effect of vascular endothelial growth factor (VEGF) presentation within fibrin matrices on endothelial cell branching. Biomaterials. 2011;32(30):7432-43.
Online Resources
Online Resource 1
Gel permeation chromatograms for L-PGA and star-PGA
Size exclusion chromatography was used to determine the dispersities (ƉM) and molecular
weights of the L-PGA and star-PGA polypeptides. SEC was conducted in 1,1,1,3,3,3-
Hexafluoro-2-propanol (HFiP) using an PSS SECurity GPC system equipped with a PFG 7 µm 8
× 50 mm pre-column, a PSS 100 Å, 7 µm 8 × 300 mm and a PSS 1000 Å, 7 µm 8 × 300 mm column
in series and a differential refractive index detector at a flow rate of 1.0 ml min-1. The systems
were calibrated against Agilent Easi-Vial linear poly(methyl methacrylate) (PMMA)
standards and analysed by the software package PSS winGPC UniChrom.
(a) (b)
S1 (a) Gel permeation chromatography (GPC) output for Benzylamine (BA)-Poly-benzyl-L-glutamate. Deprotection of this molecule results in the formation of L-PGA (b) GPC output for G2(8)-PBLG40. Deprotection of this molecule results in the formation of star-PGA.
Online Resource 2
Determination of optimal complexation procedure
Complexation time and temperature were varied to identify the optimal complexation
procedure. Particles complexed in the fridge had a larger Z-average size than those complexed
at room temperature. 5 minutes complexation was determined to be a more practical
complexation time for particle manufacture and nano-sized complexes formed in this time
period. Thus, 5 minutes complexation at room temperature was defined as the optimal
complexation procedure. Star-PGA:VEGF ratios of 10:1 or 20:1 either did not form particles
reproducibly or formed particles with a large PDI. Thus, these were omitted from further
studies. Star-PGA:VEGF ratios of 100:1 and 200:1 did not have a significantly better PDI than
the star-PGA-VEGF 50:1 formulations, thus these were also omitted from further studies as
there was no justification for the use of the higher star-PGA dose in these formulations.
Mimicking the bracketing procedure often used in the pharmaceutical industry the star-PGA-
VEGF 30:1 and 50:1 formulations were chosen for further studies.
S2 Z-average size (Z-ave), Zeta potential (ZP) and polydispersity index (PDI) of star-PGA-VEGF dispersions at various star-PGA:VEGF ratios and complexation conditions. n=3.
Star- Complexation time: Complexation time:
PGA:VEGF 5 minutes 30 minutes
Ratio
Z-ave (nm) ZP (meV) PDI Z-ave (nm) ZP (meV) PDI
Complexation in the Fridge
10:1 1796 -2.7 0.6
20:1 690.8 -4 0.6 710 -9.7 0.4
30:1 741.2 -3.2 0.4 444.5 -1.7 0.3
40:1 749.1 -4.7 0.3 399.6 -2.8 0.3
50:1 428.1 -2.1 0.3 639.1 -3.8 0.6
100:1 318.8 -2.7 0.6 422 -3 0.4
200:1 360.4 -4.9 0.3 382.9 -5.3 0.5
Complexation at Room Temperature
10:1 828.9 -3.6 0.4 363.2 -1.5 0.4
20:1 545.6 -5.4 0.5 601.3 -3.9 0.3
30:1 444.5 -2.3 0.3 568.1 -2.6 0.4
40:1 505.5 -3.2 0.3 835 -2.3 0.3
50:1 415.5 -3.6 0.2 433.6 -2.4 0.3
100:1 333.7 -2.9 0.4 407.7 -10.1 0.3
200:1 417.6 -5 0.5 334.5 -4.6 0.4
S3 Z-average size (Z-ave), Zeta potential (ZP) and polydispersity index (PDI) of L-PGA-VEGF dispersions. The L-PGA:VEGF ratio corresponds to the star-PGA:VEGF ratio with the same amount of glutamic acid units:VEGF. n=3.
L-PGA-VEGF ratio 5 minutes complexation at room temperature
Z-ave (nm) ZP (meV) PDI
20:1 539.1 -2.1 0.6
30:1 656.7 -7.7 0.4
40:1 660.9 -7.5 0.5
50:1 406.2 -7.9 0.4
100:1 753.4 -3.9 0.3
200:1 352.7 -8.8 0.3
Online Resource 3
Biocompatibility of developed star-PGA-VEGF and L-PGA-VEGF nanomedicines
Toxicity of the nanomedicines was tested to ensure their suitability for biomedical applications.
Live/Dead Cell Viability Staining was used to examine the biocompatibility of the formed
nanoparticles. HUVECs at P4 were seeded in a 24 well Corning® Costar® tissue culture
plate at a seeding density of 3 x 104 cells per well. Cells were given fully supplemented
endothelial medium for 24 hours. This medium was then removed and replaced with either
medium containing all supplements except VEGF (cells alone group), medium containing non-
encapsulated VEGF at a concentration of 50 ng/ml or medium containing no non-
encapsulated VEGF but containing the PGA-VEGF nanomedicines (VEGF 50 ng/ml). At the
selected time points: 4, 24, 48 and 72 hours, the medium was removed and the cells were
washed three times with PBS. 100 μl of live/dead solution (2 μM calcein AM, 4 μM ethidium
homodimer) was then added to the well and left to incubate for 15 minutes protected from
light at room temperature. Wells were imaged using a fluorescent microscope (Leica
Microsystems, Switzerland) and Live and Dead images were merged using ImageJ software.
As a further assessment of the toxicity of the linear and star-PGA-VEGF nanomedicines
metabolic activity of HUVECs exposed to linear PGA-VEGF 30:1, star-PGA-VEGF 30:1 or star-
PGA-VEGF 50:1 was measured. The procedure for the experiment was the same as that
outlined above except that at the selected time points of 4, 24, 48 and 72 hours, medium was
removed and 100 μl of CellTiter 96® Aqueous One Solution Cell Proliferation Assay (MTS) was
added to each well. The plate was incubated at 37°C for three hours. Absorbance was
measured on a Varioskan plate reader at 490 nm. Metabolic activity in each of the treatment
groups was normalised to that of the cells alone group at each time point.
At all time points cells exposed to VEGF free medium or medium supplemented with 50 ng/ml
non-encapsulated VEGF appeared to be alive and growing. The most dead cells observed were
in the wells containing L-PGA-VEGF 30:1 nanomedicines and cell number also appeared
lowest in this group.
Corresponding to the data obtained with the Live/Dead staining the star-PGA:VEGF 30:1 and
50:1 nanomedicines appeared to be biocompatible. The L-PGA-VEGF 30:1 nanomedicines
showed significant toxicity at the 4 hour and 72 hour time points.
(a)
(b)
to c
ells a
lon
e
Ac
tiv
ity
co
mp
are
d %
Me
tab
oli
c
150 Cells Alone
Non-encapsulated VEGF
100 L-PGA-VEGF 30:1 Star-PGA-VEGF 30:1
** ** Star-PGA-VEGF 50:1
50
0 4 Hours 24 Hours 48 Hours 72 Hours
S4 (a) Images obtained from Live/Dead staining of HUVECs following treatment with PGA-VEGF nanomedicines (b) Metabolic activity of cells in the presence of PGA-VEGF nanomedicines compared to that of untreated, healthy cells. VEGF concentration in each group: 50 ng/ml. n=3; **p<0.01.